build-with-quality
/build-with-quality
Implementuje funkcje z testami i bramkami jakości, debuguje trudne błędy i testuje projekty pod obciążeniem: TDD i EDD, uparte błędy w wielu funkcjach, przesłuchanie projektu przed kodowaniem i przegląd pokrycia, bezpieczeństwa oraz dostępności.
- Autor
- dreamlab-ai ↗
- Repozytorium
- agentbox ↗
- Licencja
- AGPL-3.0
- Ostatnia zmiana repozytorium
- 27.10.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 16
- Skrypty
- 2
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 2
Brak sygnałów. Repozytorium zmienione 27.10.2026, bez odwołań do wycofanych modeli i API.
AGPL-3.0, copyleft: przeróbki udostępniane innym muszą mieć tę samą licencję.
- 2 skrypty w paczce. Przejrzyj je przed instalacją, bo Claude może je uruchomić.
Skrypty: scripts/deepsec-gate.sh, scripts/deepsec-gate.test.mjs
Co robi i kiedy
Implementuje funkcje z testami i bramkami jakości, debuguje trudne błędy i testuje projekty pod obciążeniem: TDD i EDD, uparte błędy w wielu funkcjach, przesłuchanie projektu przed kodowaniem i przegląd pokrycia, bezpieczeństwa oraz dostępności.
Oryginalny opis (EN)
Implement features with tests and quality gates, debug hard bugs, and stress-test designs. Use when building a feature with TDD/EDD, chasing a stubborn multi-function bug (feedback-loop-first), interrogating a design before coding, or running a coverage/security/accessibility quality-gate pass. Supersedes agentic-qe, reasoningbank-*, and pair-programming.
Jak działa
- Pointers (progressive disclosure)
- When to use
- When not to use
- Quick start
- Model roles (ADR-2134)
- Methodology at a glance
- Security gate at a glance
- Related skills (complementary)
- License
Podgląd SKILL.md
---
name: build-with-quality
description: "Implement features with tests and quality gates, debug hard bugs, and stress-test designs. Use when building a feature with TDD/EDD, chasing a stubborn multi-function bug (feedback-loop-first), interrogating a design before coding, or running a coverage/security/accessibility quality-gate pass. Supersedes agentic-qe, reasoningbank-*, and pair-programming."
license: MIT
metadata:
version: 1.4.0
author: Claude Flow
tags: [meta-skill, development, qa, edd, tdd, bdd, adr, ddd, agents, quality-gates, evidence, sona, hnsw, coverage, security, accessibility, chaos-testing]
mcp_server: false
supersedes: [agentic-qe, reasoningbank-intelligence, reasoningbank-agentdb, pair-programming]
---
# Build with Quality — Unified Meta-Skill
A development + quality-engineering meta-skill: implement features with EDD/TDD,
debug hard bugs feedback-loop-first, stress-test designs before coding, and run a
coverage/security/accessibility/chaos quality-gate pipeline. Combines
[Claude Flow V3](https://github.com/ruvnet/claude-flow/tree/main/v3) with
[Agentic QE](https://github.com/proffesor-for-testing/agentic-qe) — one cohesive
system in place of several specialized skills.
## Pointers (progressive disclosure)
- **Activation prompt (start here):** [BUILD-WITH-QUALITY-PROMPT.md](./BUILD-WITH-QUALITY-PROMPT.md) — copy-paste to spin up.
- **Worked examples:** [USAGE-EXAMPLES.md](./USAGE-EXAMPLES.md) — 5 complete project examples.
- **EDD loop:** [EDD-PROTOCOL.md](./EDD-PROTOCOL.md) — Expectation-Driven Development, evidence categories, anti-fox separation.
- **Debugging:** [DEBUGGING-PROTOCOL.md](./DEBUGGING-PROTOCOL.md) — feedback-loop-first protocol, design interrogation, **Diagram-Driven Diagnosis** (complex multi-function bug / suspected parallel implementations), and **Reasoning Without a Runtime (Static-Oracle Mode)** (implementing from a spec or restoring a stub with no shell/compiler/test runner).
- **Agents catalog:** [references/agents.md](./references/agents.md) — 34 agents by domain.
- **Methodologies:** [references/methodologies.md](./references/methodologies.md) — DDD, ADR (+ ruflo ADR tooling), TDD.
- **Quality gates & workflow:** [references/quality-gates-and-workflow.md](./references/quality-gates-and-workflow.md) — gate thresholds and the 5-phase flow.
- **Review findings intake:** [references/quality-gates-and-workflow.md](./references/quality-gates-and-workflow.md#review-findings-intake) — `sealmap-review` findings (external Gemini lenses, inline fidelity) are hypotheses: reproduce with a failing test first, then fix, correct the corpus, or link the register.
- **Security gate (deepsec):** [references/deepsec-security-gate.md](./references/deepsec-security-gate.md) — run `scripts/deepsec-gate.sh --diff origin/main`; policy from `[security.deepsec]`, receipts in `.deepsec-gate/reports/`. ADR-2033.
- **Architecture:** [references/architecture.md](./references/architecture.md) — learning system, memory, model roles, consensus, MCP/CLI execution, config.
- **BHIL traceability:** [references/bhil-traceability.md](./references/bhil-traceability.md) — PRD→SPEC→ADR→TASK artifact chain, AI-native ADR types, eval/guardrail specs.
- **Migration:** [references/migration.md](./references/migration.md) — moving off agentic-qe / reasoningbank-* / pair-programming.
- **Empirical tuning:** to optimize any part of this skill against a measurable reward rather than by intuition, use the `skill-tuning` skill (SkillOpt loop + held-out A/B).
## When to use
- Implementing a feature that warrants tests and quality gates (TDD/EDD).
- Chasing a stubborn, multi-function bug — use the feedback-loop-first protocol.
- Interrogating a design before writing code.
- Running a coverage / security / accessibility / chaos quality-gate pass.
- Reviewing changed code for vulnerabilities before merge (`scripts/deepsec-gate.sh --diff <base>`).
## When not to use
- A quick code change that does not need quality gates or swarm coordination — edit files directly with Claude Code.
- GitHub-specific PR review without the full quality pipeline — use `github-code-review`.
- Standalone documentation validation and alignment — use `docs-alignment`.
- SPARC methodology orchestration without quality-engineering agents — use `sparc-methodology`.
- Simple unit test generation without the full agent system — write tests directly or use the TDD workflow in `sparc-methodology`.
## Quick start
**Option 1 — Copy-paste prompt (recommended).** Take the prompt from
[BUILD-WITH-QUALITY-PROMPT.md](./BUILD-WITH-QUALITY-PROMPT.md) and fill in:
```markdown
Build with Quality skill.
Project: [NAME] | Stack: [TECH] | Task: [DESCRIPTION]
Methodology: DDD + ADR + TDD (EDD design-time first)
Quality: 85% coverage, security scan, WCAG AA
```
**Option 2 — Agent tool + advisor (Claude Code).** Delegate to registered
subagents, choosing each one's model for the job; the lead consults the advisor
at the checkpoints in [Model roles](#model-roles-adr-2134):
```javascript
Agent({ description: "Find auth call sites", prompt: "List every caller of verify_token, very thorough", subagent_type: "Explore" }) // Haiku
Agent({ description: "Implement JWT auth", prompt: "Implement user authentication with JWT, following TDD", subagent_type: "rust-engineer", model: "sonnet" })
Agent({ description: "Auth module tests", prompt: "Write tests for the auth module to the coverage gate", subagent_type: "test-engineer" })
```
For a fan-out of many workers with a fixed shape (review N dimensions, verify
each finding), use the Workflow tool when the user has opted into it.
**Option 3 — ruflo coordination state (MCP or CLI).** `swarm_init`/`agent_spawn`
record topology and agents for observers (`/ruflo-swarm:watch`) and shared
state; they do not run Claude Code workers. On Codex, this is the coordination
surface; there the workers are Codex itself.
```bash
claude-flow swarm init --topology hierarchical --v3-mode
claude-flow agent spawn --type coder --name primary-developer
```
MCP is preferred when `mcp__claude-flow__*` tools are available; otherwise fall back
to the `claude-flow` CLI (baked on PATH — never `npx claude-flow@alpha`, which
resolves a stale cached version). See
[references/architecture.md](./references/architecture.md) for the detection logic
and the per-agent coordination hooks.
## Model roles (ADR-2134)
Claude Code's own seams decide which model runs, not ruflo:
- **Lead session:** Sonnet on high effort (`[claude_code].model`/`effort`) writes diffs and runs the suites.
- **Advisor:** Opus on call (`advisor_model`), a server-side tool Claude consults with the full transcript. Checkpoints: before a multi-file plan, schema or auth contract locks; when the same test or compiler error fails twice; before declaring done or committing, over the full diff.
- **Subagents:** pick the model per call with `model` on the Agent tool, Haiku for lookup and docs retrieval, Sonnet for implementation, Opus for design or review. Search fan-out goes to the registered `Explore` agent, pinned to Haiku.
- **Anti-fox is cross-vendor.** Haiku, Sonnet, Opus and the advisor are one model family; the evidence auditor runs on Codex (`codex exec -m gpt-6-astra`). A Claude auditor is a degraded same-family audit and must say so.
ruflo's `agent_spawn` only records an agent with a model *label*; the label runs only through ruflo `agent_execute`, a direct API call outside Claude Code. Use ruflo swarm tools for shared coordination state, not as the worker pool.
## Methodology at a glance
- **EDD** — the design-time conversation layer between human intent and AI
implementation. Captures qualitative expectations, ordering invariants, systemic
properties, and explicit counter-examples ("must NOT happen") that TDD assertions
and BDD templates can't. 7-step loop: formulate → implement → produce evidence →
audit → challenge → iterate → stabilize. Executed evidence required (command, raw
output, timestamp, git SHA); narrative evidence is auto-rejected. `evidence-producer`
and `evidence-auditor` must be different agents on different model families. Full
playbook: [EDD-PROTOCOL.md](./EDD-PROTOCOL.md). EDD runs **before** TDD/BDD, then
hands proven scenarios off to them for permanent regression coverage — it does not
replace them.
- **DDD / ADR / TDD** — bounded contexts and tactical patterns; graph-backed ADRs
with `depends_on`/`amends`/`supersedes` **plus a verification/staleness contract**
(`verified_commit` + `verified_paths` so CI catches a record drifting from its
code — an ADR you can't falsify against the code is documentation, not a
control); Red-Green-Refactor with TDD agents that also stabilize proven EDD
expectations into tests. See
[references/methodologies.md](./references/methodologies.md).
## Security gate at a glance
The Security gate is executed, not asserted: [`scripts/deepsec-gate.sh`](./scripts/deepsec-gate.sh)
runs deepsec in PR mode on the changed files, applies the manifest's
`[security.deepsec].fail_on` threshold and writes `receipt.json`. Exit `78` means
the gate is disabled or the CLI is not baked — record it as SKIPPED, never as
passed. See [references/deepsec-security-gate.md](./references/deepsec-security-gate.md).
## Related skills (complementary)
These work **alongside** build-with-quality:
- `sparc-methodology` — higher-level orchestration framework
- `swarm-orchestration` / `swarm-advanced` — lower-level swarm primitives
- `verification-quality` — truth scoring and rollback
- `github-*` — GitHub-specific integrations
- `hive-mind-advanced` — specialized Byzantine consensus features
## License
MIT
Pliki w paczce (16)
- BUILD-WITH-QUALITY-PROMPT.md 28 KB
- DEBUGGING-PROTOCOL.md 10 KB
- EDD-PROTOCOL.md 19 KB
- README.md 19 KB
- SKILL.md 10 KB
- USAGE-EXAMPLES.md 17 KB
- config/skill.yaml 15 KB
- references/agents.md 3 KB
- references/architecture.md 5 KB
- references/bhil-traceability.md 6 KB
- references/deepsec-security-gate.md 4 KB
- references/methodologies.md 5 KB
- references/migration.md 1 KB
- references/quality-gates-and-workflow.md 6 KB
- scripts/deepsec-gate.sh 15 KB
- scripts/deepsec-gate.test.mjs 10 KB
Wersje
W bazie nie ma innych skilli o nazwie build-with-quality.
Migawka bazy z 13.11.2026 · wersja paczki 7f2dbfc1b387 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 95implementation
Implementuje funkcje wraz z kodem, testami i dokumentacją według zatwierdzonych projektów, zgodnie z TDD i standardami kodowania projektu. - 100tdd
Prowadzi przepływ pracy TDD: testy najpierw, pokrycie testami i testy funkcji przed implementacją. - 100edd
Stosuje lekki TDD dla kodu pisanego z AI: przed każdym kawałkiem formułuje oczekiwania, uruchamia je i zapisuje wyniki w pliku EXPECT.md. - 100tdd
Pomaga budować funkcje i poprawiać błędy test-first, z testami integracyjnymi, w cyklu czerwony, zielony, refaktoryzacja. - 100tdd-workflow
Wymusza TDD przy nowych funkcjach i poprawkach błędów z pokryciem 90 proc. linii i 70 proc. gałęzi, w tym testami E2E. - 100tdd-implement
Implementuje funkcję lub poprawkę błędu metodą TDD: analizuje wymaganie, pisze testy, minimalny kod i refaktoryzuje.